This project involves a comprehensive approach toward the design and implementation of a therapy that will allow for solid organ allotransplants to be performed with a reduced requirement for maintenance immunosuppression. Therapies are evaluated in non-human primates (NHPs) for translation into clinical trials. The primary pre-clinical model is the rhesus monkey kidney (or in some cases skin) transplant model. Hypothesis tested. Treatment with biologicals designed to interrupt costimulatory pathways at the time of primary alloantigen exposure will prevent acute allograft rejection and induce durable allospecific tolerance. Problem and Significance. Several rodent studies have suggested that interruption of many T-cell costimulatory pathways induces a sustained state of allograft acceptance. Lasting, donor-specific allograft acceptance has been achieved without recipient morbidity using brief perioperative treatments with antibodies specific for molecules in the CD28-B7 and CD40-CD154 pathways. If this were true in humans, costimulation-based treatments would greatly reduce the expense and morbidity of transplantation. However, rodent models are overly permissive and infrequently reflect the clinical applicability of novel therapies. Furthermore, human-specific biologics must typically be evaluated in cross-reactive primate models prior to use in humans. These studies thus test the efficacy of costimulation-specific therapies in models that determine the generalizability of rodent work, and thus support translation to humans when appropriate. We have built upon previous work in this model with costimulation-blockade to demonstrate that treatment with the humanized CD154 specific monoclonal antibody (Mab) hu5c8 prevents renal and skin allograft rejection and has limited utility in reversing established rejection in NHPs. These data supported the IND application for hu5c8 facilitating the first costimulation-based transplant trials in humans. However, these initial trials performed at the NIH and elsewhere were disappointing. We observed early rejection and thromboembolic complications that were not anticipated by the rodent or NHP models. We therefore initiated studies using an alternative humanized CD154-specific Mab, IDEC-131, to determine acceptable adjuvant therapies to facilitate the reintroduction of anti-CD154 into the clinic. Since hu5c8, but not IDEC-131, essentially always prevented NHP renal allograft rejection, we opted to move to a more rigorous skin allograft model to test standard immunosuppressants for potential synergy. We have demonstrated that many conventional immunosuppressive drugs (calcineurin inhibitors, steroids, anti-IL-2 receptor antibodies) appear to hinder (or certainly do not improve) the in vivo effectiveness of anti-CD154. Conversely, the antiproliferative drug rapamycin improves the effect of IDEC-131. Furthermore, graft survival appears to be further improved by pre-transplant donor-specific blood transfusion (DST). These data are consistent with rodent data demonstrating an important role for uninhibited T-cell receptor (TCR) engagement and activation induced cell death in costimulation-induced tolerance. Our NHP skin grafts survived when sirolimus and IDEC-131 were used together (>100d vs. 8d for controls), but rejection ensued after drug withdrawal. Thus, the combination was immunosuppressive, but did not induce tolerance. When this combined therapy is used in a more permissive and clinically relevant renal allograft model, indefinite survival has been achieved in 4 of 5 animals. Animals achieving apparent tolerance have been re-challenged with donor and third party skin grafts 1 year after drug withdrawal and have accepted donor-derived graft while rejecting third party grafts, thus demonstrating the donor specific hyporesponsiveness desired for tolerance. This suggests that the combination of a brief course of IDEC-131, rapamycin, and DST may induce tolerance in some settings. A clinical trial has been designed for renal transplantation, but is now hindered by continued concerns of thrombogenicity. Necropsy evaluation of our NHPs has not shown evidence of IDEC-131-induced thrombosis suggesting either a deficiency of the model or relative safety of the compound. We have also investigated Mabs specific for the B-7 molecules CD80 and CD86 showing that they are most effective when humanized and used in combination. They do not induce tolerance in NHPs nor do they add significantly to the graft survival of hu5c8. They do appear to more effectively inhibit alloantibody formation than does hu5c8. We have determined that synergy previously suggested by earlier work between B7- and CD154-specific therapies was apparent only when the anti-CD154-specific therapy was dosed sub-optimally. Thus, although rodent data suggest that the pathways should be targeted together, the NHP data are not compelling for implementing combined therapy. Our anti-B7 data supported an IND application for the anti-B7 Mabs. Clinical trials were initiated by Wyeth but canceled for largely business concerns. At present, no forms of any Mab specific for costimulatory pathways are available for human tolerance investigation. The anti-B7 fusion protein LEA-29Y is in trials elsewhere investigating its role as an immunosuppressant. Our clinical trials remain focused on tolerance and are thus pursuing other methods, namely depletional induction. To support the depletional trials we have developed a thymectimized primate model to evaluate the role of thymic function in T-cell repopulation following therapeutic depletion. Studies are underway to evaluate animals depleted with rabbit anti-thymocyte globulin using polychromatic flow cytometry.